Factors influencing the propensity to cycle to work
Section snippets
Introduction and objectives
Cycling is widely recognised as an environmentally friendly and healthy mode of transport. However, cycling has been in long term decline in most developed countries for many years. In Great Britain, cycle traffic declined from 23 to 4.5 billion passenger kilometres between 1952 and 2003, though there is some evidence of a slight increase since the late 1990s (Department for Transport, 2004a). Between 1992 and 2002 the number of trips per person made by bicycle fell by 20% and the average
Data sources and collection
The analysis has been based on four large and complementary data sets. They cover trips to work made at least twice a week and, given the interest in competition between cycling and other modes, only journeys of miles (12 km) or less were considered to be of interest.
The National Travel Survey (NTS) provides the largest amount of choice data used here and a firm basis in actual behaviour. It contains details on individuals’ travel patterns recorded in a 7 day travel diary and was available
Mode choice model
A hierarchical logit model was developed to simultaneously estimate parameters to the two forms of RP mode choice data and the data from the two SP mode choice exercises within a single model whilst allowing for the scale of the parameters to differ between the different data sets (Bradley and Daly, 1991). The parameters are estimated in units of residual variation of the upper nest and hence the data set expected to have the most appropriate residual variation for forecasting should be placed
Forecasting applications
Table 2 indicates the 1997 commuter market shares in the NTS data. Amongst those who commute miles or less, 62% have a car available for their journey to work. The vast majority of these use it and the cycle share is very minor. As would be expected, cycle picks up a much larger share of those who do not have a car available, although its share of the non car market is similar. To serve as a contrast, the figures for commuting over miles are given and cycle has a negligible share,
Conclusions
This study has developed what we believe to be the most comprehensive and largest model which handles cycling within mode choice, at least within the British context. The model can be used to forecast, at either a strategic or more detailed level, urban mode choice under a variety of scenarios, with particular emphasis on cycling which is a neglected area of research but which is expected to become increasingly important because of the recognition of its potential in terms of health promotion
Acknowledgement
The authors would like to thank the Economic and Social Research Council who sponsored the research (Grant R000237103) and the numerous cyclists and non-cyclists who agreed to take part.
References (29)
- et al.
Cycling to work from wards in a metropolitan area. 3 Testing the model
Traffic Engineering and Control
(1989) - Booost, U.K., 2005. Provide your employees with a bike for 50% less than retail prices. www.booost.uk.com (accessed...
- et al.
Route choice analysed with stated preference approaches
Transportation Research Record
(1985) - Bradley, M.A., Daly, A.J., 1991. Estimation of logit choice models using mixed stated preference and revealed...
- Crespo Diu, F., 2000. Cycling to work in Great Britain: a quantitative analysis. MA Dissertation, Institute for...
- Davies, D.G., Halliday, M.E., Mayes, M., Pocock, R.L., 1997. Attitudes to cycling: a qualitative study and conceptual...
- Department of Transport, 1996. The National Cycling Strategy. HMSO,...
- Department for Transport, 2004a. Transport Statistics Great Britain 2004 Edition, London,...
- Department for Transport, 2004b. The Future of Transport a Network for 2030 Cm6234....
- Department for Transport, 2005. Focus on Personal Travel 2005 Edition, London,...